心脏MRI应变预测心脏结节病PET心肌炎症
本研究回顾性纳入125例符合日本循环学会标准的确诊或很可能心脏结节病患者,均在90天内完成心脏MRI和FDG PET/CT。通过特征追踪计算应变参数,分析其与PET/CT定义的局灶性心肌炎症的关系。结果显示圆周应变和纵向应变在炎性节段显著降低,并独立预测PET定义的心肌炎症。
要点
- 心脏结节病中PET炎性节段的圆周应变和纵向应变绝对值均低于非炎性节段。
- 圆周应变加入左室射血分数后可提升对PET心肌炎症的检测效能。
- 应变分析可作为心脏结节病心肌炎症的辅助影像标志物。
核心结论
- 心脏MRI的圆周应变和纵向应变独立预测PET/CT定义的心肌炎症。
研究设计与样本
- 回顾性研究,纳入125例确诊或很可能心脏结节病患者,均在90天内完成心脏MRI和FDG PET/CT。
- 依据日本循环学会标准,采用17节段美国心脏协会模型,以局灶性FDG摄取定义炎症。
- 应变参数通过电影稳态自由进动序列图像特征追踪计算,包括峰值圆周应变、峰值纵向应变和舒张期应变率。
关键影像表现/征象
- 炎性节段的峰值圆周应变低于非炎性节段(-14.1% vs -18.4%,P < .001)。
- 炎性节段的峰值纵向应变低于非炎性节段(-10.3% vs -13.0%,P < .001)。
- 18.1%(385/2125)的节段在FDG PET/CT图像上显示心肌炎症。
诊断标准与关键数值
- 多变量分析中峰值圆周应变和峰值纵向应变均独立预测炎症(比值比分别为1.79和1.62,均P < .01)。
- 将峰值圆周应变加入左室射血分数后,炎症检测的ROC曲线下面积从0.63提升至0.72(P = .04)。
- 多变量逻辑回归校正了晚期钆增强、T2信号异常、年龄、性别和左室舒张末期容积指数。
临床应用与阅片要点
- 对于疑似心脏结节病心肌炎症的患者,心脏MRI应变参数可作为FDG PET/CT之外的补充评估手段。
- 应变绝对值降低的节段应重点关注有无活动性心肌炎症,结合PET/CT局灶性FDG摄取综合判断。
局限性
- 回顾性单中心研究设计。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
Purpose To determine whether cardiac MRI strain parameters independently predict PET/CT-defined myocardial inflammation in patients with cardiac sarcoidosis (CS). Materials and Methods This retrospective study included patients with definite or probable CS who underwent cardiac MRI and fluorine 18 fluorodeoxyglucose (FDG) PET/CT between January 2010 and October 2022 within 90 days per the Japanese Circulation Society criteria. Strain parameters, including the peak circumferential strain (PCS), peak longitudinal strain (PLS), and diastolic strain rates, were calculated using feature tracking on cine steady-state free precession images. Inflammation was defined by focal FDG uptake using the 17-segment American Heart Association model. T2-weighted and late gadolinium enhancement (LGE) images were qualitatively assessed. Segment-level comparisons were performed between inflamed and noninflamed segments. Multivariable logistic regression clustered at the patient level was adjusted for LGE, T2 signal abnormality, age, sex, and left ventricular end-diastolic volume index. Receiver operating characteristic (ROC) curve analysis was used to evaluate the added value of strain metrics over the left ventricular ejection fraction (LVEF) for inflammation. Results Among 125 included patients (mean age, 59.5 years ± 12.7 [SD]), 60% (n = 75) had at least one segment with inflammation on FDG PET/CT images. Of these 75 patients, 61% (n = 46) were male. Segment-level analysis revealed inflammation in 18.1% (385 of 2125) of the segments. The PCS and PLS were lower in inflamed segments than in noninflamed segments (-14.1% vs -18.4% and -10.3% vs -13.0%, respectively; P < .001 for both) and independently predicted inflammation in multivariable analysis (odds ratio, 1.79 and 1.62, respectively; P < .01 for both). Adding PCS to LVEF improved inflammation detection (area under the ROC curve, 0.72 vs 0.63; P = .04). Conclusion Circumferential and longitudinal strain from cardiac MRI independently predicted PET-defined myocardial inflammation. Keywords: MRI, PET/CT, Heart, Cardiomyopathy, Cardiac Strain, Cardiac Sarcoidosis, Myocardial Inflammation © RSNA, 2026.
引用
- Firoozeh, Shafiei, Gulhane, et al. Cardiac MRI Strain for Prediction of PET-defined Myocardial Inflammation in Patients with Cardiac Sarcoidosis. Radiol Cardiothorac Imaging. 2026. PMID: 42690155. doi:10.1148/ryct.250318
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。